探索 | Thread Easy - Twitterスレッドを展開 | リーダー・要約・作成

$Of course DeepSeek has caching at home. At pass@5 70.2 they are still at Opus 4.5 level at a tiny fraction of the cost, and only really lose to GPT 5.2 and well-designed tool systems. This should give you an idea of the next gen's pass@1 ceiling.$

Of course DeepSeek has caching at home. At pass@5 70.2 they are still at Opus 4.5 level at a tiny fraction of the cost, and only really lose to GPT 5.2 and well-designed tool systems. This should give you an idea of the next gen's pass@1 ceiling.

We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1

Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)

Wed Dec 17 21:32:40

Mans literally became a meme on X and is now working at XAI. Elon you’ve made a great choice in the simulation

stealth consumer bio | prev founding team @a16z @speedrun | writer of the bird’s nest, https://t.co/3fgQ6avUsx

robin

Wed Dec 17 21:29:45

Gradient descent is a powerful tool for optimization spaces that verify the manifold hypothesis, but the space of reasoning is discrete and combinatorial. GD fails in cliff-like landscapes where a single discrete change (a logical step) alters the entire outcome. Unless...???

Co-founder @ndea. Co-founder @arcprize. Creator of Keras and ARC-AGI. Author of 'Deep Learning with Python'.

François Chollet

Wed Dec 17 21:26:13

Gradient descent is a powerful tool for optimization spaces that verify the manifold hypothesis, but the space of reasoning is discrete and combinatorial. GD fails in cliff-like landscapes where a single discrete change (a logical step) alters the entire outcome. Unless...???

Co-founder @ndea. Co-founder @arcprize. Creator of Keras and ARC-AGI. Author of 'Deep Learning with Python'.

François Chollet

Wed Dec 17 21:26:13

RT @AngelLamuno: Still, there's always a tension between the biological roots in the biological organism of the human psyche and the transc…

Co-founder @ndea. Co-founder @arcprize. Creator of Keras and ARC-AGI. Author of 'Deep Learning with Python'.

François Chollet

Wed Dec 17 21:25:26

RT @yatharthmaan: Larry Page: “Elon is one of the people who genuinely understands physics and applies it.” Sergey Brin: “Elon is someone…

🏗️ Love to build (@runwayco @sandboxvr @zynga) people love 💸 Investor @amplitude_hq @mercury @owner @elevenlabsio @meetgamma @sfcompute @turingcom++

Siqi Chen

Wed Dec 17 21:18:56

探索

Newest first — browse tweet threads

探索

Newest first — browse tweet threads

Of course DeepSeek has caching at home. At pass@5 70.2 they are still at Opus 4.5 level at a tiny fraction of the cost, and only really lose to GPT 5.2 and well-designed tool systems. This should give you an idea of the next gen's pass@1 ceiling.

Mans literally became a meme on X and is now working at XAI. Elon you’ve made a great choice in the simulation

Gradient descent is a powerful tool for optimization spaces that verify the manifold hypothesis, but the space of reasoning is discrete and combinatorial. GD fails in cliff-like landscapes where a single discrete change (a logical step) alters the entire outcome. Unless...???

Gradient descent is a powerful tool for optimization spaces that verify the manifold hypothesis, but the space of reasoning is discrete and combinatorial. GD fails in cliff-like landscapes where a single discrete change (a logical step) alters the entire outcome. Unless...???

RT @AngelLamuno: Still, there's always a tension between the biological roots in the biological organism of the human psyche and the transc…

RT @yatharthmaan: Larry Page: “Elon is one of the people who genuinely understands physics and applies it.” Sergey Brin: “Elon is someone…